Re: Linux & ECC memory
Rob Hagopian (Rob.Hagopian@vuser.vu.union.edu)
Sun, 17 Nov 1996 19:12:18 -0500
>> 1) Couldn't this code determine the active process and remove the active
>> memory pages from use? (although you'd definatly want this as an option...
>> I see that some powersaving modes also generate an NMI!)
>
>The NMI may occur asynchronously to the causing process. If your L1 cache
>or L2 cache is write back then the cache write back could occur seconds
>after (in theory even longer) the write to the bad address.
Mmmm... yes, that's right... oh well, I sort of reconsidered this idea
later, the kernel bloat would probably be unjustified anyways. <sigh>
Now for the P6 and/or ECC memory where there's more reporting as to where
the memory fault occurred, this might be "a good thing".
-Rob